Image Encryption Scheme by Applying Heterogeneous Chaotic Neural Network: A Review Article

Abstract

This paper introduces a new combined pseudorandom sequence generator based on neural network and chaos and a chaotic encryption algorithm based on DNA rules for safe transmission and storage of images. A new heterogeneous chaotic neural network generator that governs the operations of the encryption algorithm is used in the proposed scheme: pixel location permutation, DNA-based bit substitution and a new form of DNA-based bit permutation. By dynamically updating control parameters as well as the number of iterations of the chaotic functions in the neural network, the randomness of the generated chaotic sequence is improved. Many experiments are conducted to demonstrate a high degree of randomness of the proposed chaotic generator, including auto correlation, 0/1 balance and NIST tests. To demonstrate the safety and efficiency of the proposed chaos-based genetic encryption process, experimental results such as pixel correlation coefficients, entropy, NPCR and UACI, etc., as well as security analyses are provided.